Lithium niobate, LiNbO.sub.3, is well known for its electrooptic, ferroelectric and piezoelectric properties, and is widely used in integrated and guided-wave optics applications, such as couplers, switches, modulators, deflectors and rf spectrum analyzers.
Although polycrystalline and single crystal LiNbO.sub.3 have been prepared by a variety of standard solid state synthetic methods, relatively little work has been reported dealing with the fabrication of thin films of this material. Magnetron and of sputtering and molecular beam epitaxy have been used to fabricate thin films of LiNbO.sub.3. These techniques, however, require complicated and expensive high vacuum equipment. It is also difficult, using these processes, to maintain film uniformity over large areas. Further disadvantages are the low growth rates associated with these techniques as well as their difficulty in providing conformal coverage. Conformal coverage is the degree to which the deposition follows the contour of the substrate surface. Line of sight techniques such as sputtering typically exhibit poor conformal coverage.
In "Preparation of Crystalline LiNbO.sub.3 Films with Preferred Orientation by Hydrolysis of Metal Alkoxides," Advanced Ceramic Materials, Vol. 3, No. 5, 1988, a process is disclosed for preparing LiNbO.sub.3 thin films using single source reagents by a sol gel process utilizing hydrolysed metal alkoxides, such as, for example, an ethanol solution of lithium ethoxide and niobium ethoxide. It is difficult to obtain epitaxial films using sol-gel deposition techniques as it is not possible to control the crystallization rate using this process.
U.S. Pat. No. 5,051,280 discloses a method for producing alkali metal niobates and tantalates which involves pyrolyzing the stoichiometric salt of an alkali metal and a niobium or tantalum complex of a bidentate or tridentate ligand. For example, the salt may be dissolved and then coated on a substrate, after which the coating is pyrolyzed at 400.degree. C. for 10 minutes. It is difficult to obtain high quality films having monocrystallinity using such a organometallic decomposition method, due to the lack of control in the crystallization rate as well as the relatively large amount of organic material that is evolved during the processing of the coated precursor layer.
Chemical vapor deposition (CVD) is a technique capable of depositing films at high growth rates. To date there has been little work done on depositing complex metal oxide films such as LiNbO.sub.3 by CVD.
One method for forming LiNbO.sub.3 films by chemical vapor deposition is disclosed in Curtis and Brunner U.S. Pat. No. 3,911,176, and in "The Growth of Thin Films of Lithium Niobate by Chemical Vapor Deposition," by B. Curtis and H. Brunner, Mat. Res. Bull. Vol. 10, pp. 515-520, 1975. Both of these references disclose vaporizing separate lithium and niobium compounds and bringing the resultant vapors in contact with a heated substrate in an oxidizing atmosphere. Preferred materials for producing the vapor precursor include lithium chelates of beta-diketonates and niobium alkoxides. One disadvantage to a process such as this, which utilizes separate sources for each metal precursor, is that it is difficult to make each precursor volatilize at the same rate. Consequently, because there is usually more of one metal than the other in the precursor vapor, it is difficult to achieve the desired 1:1 stoichiometry in the resultant LiNbO.sub.3 film. In addition, when dual source precursors are involved, one of the metals may deposit preferentially, leading to an undesirable film composition. Further, the films deposited using the Curtis CVD method are initially black, and therefore not useful in many electro-optic applications which require high light transmittance, until they have first been annealed.
"Metal Alkoxides as Precursors for Electronic and Ceramic Materials", by Bradley, Chem. Rev., Volume 89, pp. 1317-1322 (1989), mentions that barium titanate, BaTiO.sub.3, can be deposited on a quartz substrate using the heterometal alkoxide BaTi(OPr.sup.i).sub.6 in 2-propanol solution. However, there is no indication of the method utilized to achieve this deposit.
Thus, there continues to be a need for a chemical vapor deposition process that will produce monocrystalline complex metal oxide films which have an epitaxial relationship to the substrate and, preferably, do not require subsequent annealing to produce optically transparent films. The same need holds true for other common complex metal oxides which exhibit useful electrooptic, ferroelectric and piezoelectric properties, in particular the vanadates, niobates, tantalates, zirconates, hafnates and titanates of alkali metals, alkaline earth metals, and lead.